51[01:08:33] <MikeDebian> guys, is it possible to use extundelete on raid mdadm raid array?
52[01:08:35] <MikeDebian> need to recover some unintentionally deleted files...
53[01:10:03] <MikeDebian> the array is already stopped, now Im trying to mount it as readonly, though not sure if mdadm --assemble --readonly /dev/mdX us enough. mount -o remount, ro /dev/mdX /mnt/whatever is returning "bad option" error
112[01:28:03] <MikeDebian> assuming the array is mounted in /mnt/raid and I want to restore a full directory I should run: "extundelete /dev/mdX --restore-directory /mnt/raid/home/whatever/directory/to_recover/"
113[01:28:06] <MikeDebian> right?
114[01:28:28] <ratrace> I never had to use it so I wouldn't know, sorry.
115[01:28:50] *** Quits: milkt (~debian@replaced-ip) (Remote host closed the connection)
164[02:13:52] <MikeDebian> extundelete is not giving me much luck
165[02:14:18] <MikeDebian> anyone has used ext4magic before?
166[02:14:58] <MikeDebian> "ext4magic /dev/mdX -f directory/to/recover" lists me all the files that were there
167[02:15:06] <MikeDebian> this gives me some hope
168[02:15:44] <MikeDebian> but when trying to recover the directory or the files it doesnt work. even recovering files that I know were not deleted is not working. I might be doing something wrong
178[02:25:24] <MikeDebian> the files were related to an internal nvr web software that I've been developing. In the meantime I remembered that I actually backed it up on the 30th of December so I have a backup at least. but some great improvements have been done since then. I wouldn't like to be redoing them
179[02:25:55] <ratrace> let me introduce you to my friend Git, and his friend, daily commits. :)
180[02:26:17] <ratrace> (and his friend's friend, git push to a remote origin)
193[02:33:41] <twb> ratrace: the biggest problem with rsnapshot is its stupid bloody config file
194[02:34:09] <twb> but they never gave me a budget to fix it
195[02:35:22] <dvs> ??? I didn't find it too difficult.
196[02:35:59] <ratrace> it worked for me. iirc there was some gotchas about it, but in the end, it worked mighty fine.
197[02:36:05] <qman__> I wrote my own rsync script ages ago
198[02:37:23] <qman__> I actually use backuppc for most of my stuff, only use the rsync script on a couple systems where they're backing up to other locally attached storage
199[02:38:18] <ratrace> I still use rsync, but I'm shipping zfs and btrfs snapshots around
203[02:41:53] <qman__> I'm actually using the rsync script to back up a non-ZFS volume to a ZFS volume, and then ZFS sending it to another machine
204[02:42:15] <qman__> with snapshots
205[02:43:18] <twb> rsnapshot uses literal tabs as inter-record separators, and also if you e.g. do not set cmd_cp to /bin/cp, it will instead use a vastly less efficient in-perl cp
206[02:43:27] <twb> i.e. it's not smart enough to just *look* for cp in $PATH
222[03:00:12] <edufmass> Hello, I need some advice. When installing the system a host is defined then a domain. I have a domain pointing to my home ip to use with apache, etc. I don't know the relationship between real domain and domain defined in installation
223[03:01:16] <sney> the domain field in the installer is only used for dns purposes.
224[03:01:59] <twb> if you're on a home LAN you probably don't have a real domain anyways
225[03:02:08] <sney> apache may (I don't use apache so I don't remember) auto populate some configs from that, but it doesn't have a major effect on anything, and you can change/remove it in /etc/resolv.conf and hosts
226[03:02:16] <twb> and the local stuff will all be on .local MDNS
243[03:23:29] <Mazhive> funny thing when i play dota on screen one and move my cursor to the second screen it changes the current window (screen dota ) to my main screen like you would do with alt tab
244[03:24:00] *** Quits: CombatVet (~c4@replaced-ip) (Remote host closed the connection)
267[03:39:04] <fling> jmcnaught: it gives me no results -> replaced-url
268[03:39:27] <fling> So which package is config coming from? And why is it not on the search? ^
269[03:40:03] <jmcnaught> fling: mine is from buster-backports, yours will be different if you are using the stock buster kernel. If you run "ls /boot/config-$(uname -r)" it will show you yours.
270[03:40:27] <fling> jmcnaught: and what is the package name?
272[03:41:26] <jmcnaught> fling: for me it is linux-image-5.9.0-0.bpo.2-amd64, for you with stock buster kernel it would be linux-image-4.19.0-13-amd64
336[04:20:55] <fling> The issue is stuff using openmp and similar libs crashes on this numa box, not sure if this is because of kernel or something else
338[04:21:20] <twb> So e.g. you can cp /boot/config-4.9 linux-5.0/.config and then "make syncconfig" there, and it'll only prompt you about new options
339[04:21:43] <fling> yeah
340[04:21:59] <twb> You probably also should knoe about "make deb-pkg" which is the recommended way to spit out an (unsigned) kernel
341[04:22:02] <fling> how is it different from oldconfig?
342[04:22:08] <twb> oldconfig just changed its name
380[05:04:17] <alex11> hi, having some major issues with my xfce desktop
381[05:04:57] <alex11> i ran some compiz setting and it borked things, couldn't open window manager settings; so i rebooted, removed compiz and tried to log in, and now it's in this weird state of being borked
382[05:05:47] <alex11> i wonder if having removed compiz i need to do another reboot first, maybe it loaded that first when lightdm started up
425[05:19:52] <alex11> hmm, it apparently still starts compiz
426[05:20:05] <alex11> i have to invoke 'xfwm4' to get proper gui behavior again
427[05:20:51] <alex11> what was the config file people wanted?
428[05:21:48] <JackFrost> twb recommended trying to nuke your Xfce config and try logging back in, I add in that you should also dump ~/.cache/session (or whatever it is.)
429[05:22:14] <twb> also you don't need a full reoot
430[05:22:15] <alex11> in the tty i guess
431[05:22:29] <JackFrost> Ensure xfconfd isn't running, just to make sure.
433[05:22:32] <twb> you can just restart GUI, e.g. "pkill -u alex11" or "sudo systemctl restart lightdm"
434[05:22:40] <alex11> how do i nuke my xfce config, what part?
435[05:23:11] <twb> grep -r xfwm4 ~/.config
436[05:23:14] <twb> grep -r compilz ~/.config
437[05:23:23] <twb> alex11: ^ I'd try some basic things like that to guess
438[05:24:30] <alex11> can i get it back later? i don't want to lose my customizations
439[05:24:30] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
440[05:24:54] <twb> well just move the files, don't delete em
441[05:25:24] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
442[05:27:14] <alex11> what's the playbook, here? nuke the configs, log back in and... then what?
443[05:28:20] <alex11> btw all this stuff is in /home/alex/.config/xfce4/xfconf/xfce-perchannel-xml/
444[05:28:26] <alex11> if you want the exact file path
445[05:28:56] <alex11> i don't understand why there isn't a simpler way to just 'make it work', if i removed compiz
446[05:29:09] <alex11> instead of removing configs
447[05:30:36] <JackFrost> Did you try the grep commands, like twb recommended..?
448[05:30:54] <alex11> yes, that's where i got the file path
449[05:31:21] <alex11> compiz returned nothing and xfwm4 returned /home/alex/.config/xfce4/xfconf/xfce-perchannel-xml/xfwm4.xml and /home/alex/.config/xfce4/xfconf/xfce-perchannel-xml/xfce4-keyboard-shortcuts.xml
450[05:31:28] <alex11> which i've now copied to another folder
464[05:37:58] <alex11> /usr/bin/xfwm4 for x-window-manager and "/etc/alternatives/x-session-manager: X server already running on display :0.0" and "xfce4-session: Another session manager is already running" for session-manager
465[05:38:47] <twb> that's... extremely silly
466[05:38:52] <quadrathoch2> MikeDebian hope it works out for you. If not look into photorec and testdisk (apart from the photorec name it does recover more than photos) and I hope you learned your lesson ^^
467[05:39:07] <twb> I think you accidentally put a semicolon in there
495[05:49:43] <JackFrost> alex11: I'd start out just trying to rm sessions and ensure that the checkbox isn't saved on logout.
496[05:49:45] <twb> alex11: sorry I've done all the basic stuff I know
497[05:49:59] <twb> alex11: did you ask #xfce
498[05:50:05] <alex11> no not yet
499[05:50:24] <alex11> JackFrost, i might just uncheck 'save session' and then reboot then? seems like that alone might do the trick
500[05:50:53] <alex11> or, well, log out
501[05:50:55] <alex11> not reboot
502[05:50:59] <abhixec> Hi all! I am currently running XMonad+arch on my laptop but am considering debian. I don't particularly have a need for latest linux kernel but I want certain software packages to be latest(xmonad, xmonad-contrib, emacs, vim,..etc) would debian work for me?
503[05:51:04] <twb> save session basically means it'll remember that you had firefox open, and re-start firefox on login
504[05:51:20] <twb> I don't *think* it helps with xfce internals
505[05:51:53] <alex11> JackFrost, i guess my question is, if i'm just restoring the config later on, won't it re-cause these issues?
507[05:52:02] <twb> abhixec: depending how much stuff, you can run Debian Stable and then cherry-pick from backports, or you can run Debian testing or testing/unstable -- these are essentially "rolling releases"
508[05:52:39] <JackFrost> alex11: Again, my first recommendation would simply be ~/.cache/sessions/, which would not later be restored.
509[05:52:51] <alex11> abhixec, window managers don't tend to get updated so you might want debian testing, usual caveats apply, don't run it on anything critical
510[05:53:03] <alex11> window managers don't tend to get backported *
511[05:53:24] <alex11> JackFrost, ok. let me try that. logging off hexchat so i can go back to tty
542[06:08:42] <alex11> and why i thought it was compiz
543[06:08:52] <alex11> because i knew there was 'compton'
544[06:09:09] <alex11> and i said to myself "no don't install that, you want the other one, that one is a compositor"
545[06:09:16] <abhixec> Thanks twb alex11 so I can still have a mix of stable + rolling(i read it is sid correct) for the software that I want correct?
546[06:09:25] <alex11> except instead of conky my brain went to compiz
547[06:09:38] <alex11> no, you shouldn't mix stable with testing/sid
575[06:15:07] <alex11> a lot of people who say 'i need a newer version of such and such' actually don't
576[06:15:14] <pclover> this is probably covered in the wiki but i guess i can ask anyways. does apt-get have an option to exclude packages from being installed like it exists for yum on rpm based distros?
577[06:15:19] <alex11> but yeah, some people think testing is fine, so use it if you like
578[06:15:19] <pclover> or something similar to it
579[06:15:29] <alex11> pclover, apt-mark hold, i think
580[06:15:43] <alex11> but dependencies exist for a good reason
593[06:17:29] <pclover> tho, it's probably better to build a package instead
594[06:17:51] <JackFrost> Yep, at least a dummy one that provides it.
595[06:18:08] <alex11> abhixec, i feel like you might be fine with debian testing; just keep backups and watch what the package manager tells you when installing/upgrading stuff
596[06:18:31] <alex11> don't run it on mission critical things - but that being said, it's debian, and 'in development' is as good as some other distro's 'stable' releases
597[06:18:35] <pclover> unless there is a way to install specific packages from debian-testing but that seems like a bad idea
598[06:19:17] <JackFrost> Via apt pinning, but with mixed results yeah.
603[06:21:23] <alex11> so abhixec if you do want to try compiling your own xmonad on debian stable their site does give you compile instructions
604[06:21:26] <alex11> no idea how well it will work
605[06:26:22] <abhixec> that is a good point I need to see if any of the modules that I use in config have dependency on latest xmonad/xmonad-contrib, yeah I agree I mean one of reasons I am looking at Debian is because I feel with Arch I am spending way too much time trying to get my laptop in a working state(with nvidia+ latest kernel messing up my ability to connect to external monitor)
606[06:27:14] <alex11> yeah i mean if you know for sure you need the newest xmonad it's one thing but if you're not sure you might be able to make do with stable's
607[06:27:24] <alex11> vim/emacs are more straightforward, you don't need testing/sid for that
608[06:27:51] *** Quits: akp55_ (~akp55@replaced-ip) (Remote host closed the connection)
609[06:29:29] <alex11> debian testing is fine, you just need to understand the limitations and not run too many critical things on it, because it's in development, packages can get removed, it gets security fixes slower than other releases
611[06:30:55] <abhixec> is there a guide to installing debian + xmonad I see the media ships with xfce4 correct?
612[06:31:54] <alex11> if you want no desktop to start with, just unclick the desktops in the installer and then from the shell (when it finishes installing), apt install xorg xmonad and whatever else you want
613[06:32:04] <alex11> you can start it with either a display manager or startx
616[06:34:29] <alex11> if you have an internet connection you can just use the netinst image, or the netinst with non-free firmware if you're on wifi or somethijng
617[06:34:31] <alex11> something
618[06:34:42] <alex11> to get your devices to work in the installer
619[06:35:21] <twb> Note that "startx" is unlikely to place nice with systemd-logind
620[06:35:45] <twb> xmonad and xfce will probably deal with it, but GNOME is likely to have a huge whinge
624[06:37:11] <twb> abhixec: Debian ships big (ahead-of-time) and small (on-demand) install media. We also have Debian Live images for several desktops, including XFCE, but not xmonad.
625[06:37:30] <twb> abhixec: you can install any of them any then add/change desktop or WM whenever you want
626[06:37:42] <alex11> yeah what twb said
627[06:37:47] <alex11> in a better way than i did
628[06:38:02] *** debhelper sets mode: +l 1136
629[06:38:14] <twb> Because xmonad isn't a full desktop, you can't just choose it as the preferred desktop at install time
630[06:38:27] <mcole> hello, quick question, has modprobe been removed from Debian completely? I want to shush my built-in laptop beeper but can't blacklist pcspkr via modprobe
631[06:38:55] <twb> mcole: did you try "blacklist pcspkr" in /etc/modprobe.d/die-pckspr-die.conf ?
632[06:38:56] <mcole> and when I search for it with apt, there are no results, neither in packages.debian.org
633[06:39:19] <mcole> twb: nope, but that sounds like a good start
634[06:39:24] <twb> mcole: the "kmod" package provides "modprobe"
635[06:39:26] <abhixec> thanks twb,yeah I understand that. I want to run xmonad so I guesss I will get the small(on-demand) one and try installing!
636[06:40:00] <twb> mcole: however you will likely run into two issues: 1. pcspkr might be compiled =y; and 2. "blacklist" doesn't do what you think, you generally need "install foo /bin/false"
637[06:40:12] <mcole> twb: thanks, I was wondering where to find it. Which should be my preferred method, manually editing or running the command?
638[06:40:31] <twb> mcole: also for pcspkr I vaguely recall that there's some other separate thing
720[07:37:55] *** Quits: black_ant (~antilope@replaced-ip) (Quit: simplicity does not kill)
721[07:38:42] <Logg> Wondering if any of you nerds know if there's a security risk in passing access to your X server over an SSH session? Could the remote server grab screen buffer contents even for things that it's not trying to render itself?
754[08:07:46] <Rodon> what command (systemctl reboot) works for instant reboot ?
755[08:08:05] <iamchroot> i thought reboot would do it
756[08:08:25] <tomreyn> an instant reboot is a power cycle, i assume?
757[08:09:23] <Rodon> "systemctl reboot" gives empty screen ..it takes much longer to reboot..
758[08:09:45] <Rodon> kind of freeze i see..
759[08:10:21] <JackFrost> twb: Yeah I don't like CSDs at *all*. :)
760[08:10:39] <tomreyn> this suggests something is wrong about a systemd service you are running. would need to have a look at the logs or boot with verbose output to know whats the problem.
761[08:11:09] <twb> Rodon: there are several reasons that might be
770[08:13:14] <Rodon> twb: thanks..it makes sense now..
771[08:13:34] <twb> Rodon: if this is happening every time, you probably want to work out which unit in systemd is causing the hang. This is a HUGE pain in the arse, because systemd shuts down most logging early in shutdown/reboot cycle :-(
772[08:13:39] <tomreyn> oh and systemd services failing to shut down properly can also CAUSE DATA LOSS
773[08:13:50] <twb> Rodon: the best option I found was to hook up the serial line to another computer
774[08:14:32] <twb> tomreyn: it's a different class of failure, though. systemd hangs are purely in userland. "reboot -ff" will bypass low-level shit like parking the HDD heads
775[08:14:57] <tomreyn> i see
776[08:15:04] <Rodon> i want safe reboot..
777[08:15:14] <tomreyn> To get a more verbose log of the boot *and shutdown* process, remove "quiet" and "splash" from the kernel boot parameters and add "debug systemd.log_level=info".
778[08:15:20] <twb> Rodon: oh one other thing you can try: if you type Ctrl+Alt+Del 7 times in 2 seconds, systemd will do a hard reboot
779[08:15:37] <twb> tomreyn: ah yeah thanks I forgot those are on by default :-
780[08:15:59] <tomreyn> that way you should not just get a black screen but get a better idea of where it's hanging
784[08:16:31] <twb> Also if plymouth is installed, you can hit Esc during shutdown and it'll re-print all the console text
785[08:16:53] <twb> This is assuming the hang is early enough that the keyboard hasn't been disconnected &c
786[08:17:43] <twb> There's a bunch of other Deep Magic tricks e.g. dracut initrd pivot_root during shutdown, but you don't want to think about that stuff if simpler options work
875[09:29:53] *** Quits: Ycarus (~Ycarus@replaced-ip) (Remote host closed the connection)
876[09:30:58] <FishMonkey> thanks JackFrost. i made a panel icon of $mousepad .local/share/notes/Notes/Notes # that should keep me satisfied until it comes back, lol
989[11:23:48] <Darcidride> Hi guys, when running "sudo powertop" I don't have any issues, even "sudo powertop --auto-tune" is working without any issues, but when i'm trying to execute "sudo powertop2tuned -n -e laptop" it's not working and returned me "PowerTOP returned error code: -11" which is weird, i search a bit, but after a while i only find an unresolved issue from Fedora 31 replaced-url
990[11:23:53] <Darcidride> things i can do to debug this issue ? (i don't even know what's this error -11 is :/)
994[11:29:47] <diogenes_> DarkaMaul, what is powertop2tuned and where did you get it from?
995[11:29:51] <diogenes_> v, powertop2tuned
996[11:30:15] <diogenes_> v. powertop2tuned
997[11:37:00] <misternumberone> hi, i'm restricted by driver limitations to a custom kernel based on 4.14.108. This seems to work fine in debian 10. however, i'd like to try software that's only for debian sid (replaced-url
998[11:37:32] *** Quits: tyzef (~tyzef@replaced-ip) (Remote host closed the connection)
1005[11:44:54] <misternumberone> however, the part where i thought "i can't use this software as things stand" is when i noticed it needs gcc 10. i'm pretty sure if i update that beyond what debian 10 comes with, many things may stop working, without full dist-upgrade
1007[11:45:11] <Darcidride> misternumberone, theorically if you're using apps (stable for example), if should be tested and working accordingly to the stable kernel, so you should be able to use them, for more recent software, with an old kernel/drivers, i'm not sure it should work, it could, but i don't thing the maintainers are testing the software backward compaability with old kernel (maybe i'm wrong, correct me if it's the case)
1008[11:45:39] <Darcidride> it / think *
1009[11:46:06] *** Quits: yonder (~yonder@replaced-ip) (Remote host closed the connection)
1010[11:46:16] *** Quits: ottavio (~m0ttv@replaced-ip) (Remote host closed the connection)
1092[13:18:08] <RadoS> With i5-....U CPUs inten says it has 2 cores, 4 threads, with i5-....HQ CPUs it says 4 cores, 8 threads. Yet for both /proc/cpuinfo reports 4 cores, and in top for both 4 cpus are shown. What am I misunderstanding?
1101[13:25:54] <ratrace> i5 are traditionally not HT. I think only latest gens (skylake onward or something liek that) are also HT. which one is yours?
1102[13:26:06] <ws2k3> how can i increate the heap for zookeeper on debian?
1103[13:26:12] *** Quits: CrystalMath (~coderain@replaced-ip) (Quit: Call-out culture is toxic, disgusting, and vile!)
1104[13:27:14] <rk4> ws2k3: commonly that's a jvm commandline arg
1106[13:27:34] <ws2k3> rk4: yeah i know that. but im trying to find how where is should add it
1107[13:29:37] *** Quits: dbristow (~dbristow@replaced-ip) (Remote host closed the connection)
1108[13:30:00] <jelly> ratrace, my i5-660 was HT. 2010.
1109[13:31:51] <jelly> RadoS, the machine with your i5-....HQ cpu might have HT disabled, what is the actual model, and can you pastebin complete /proc/cpuinfo
1126[13:50:39] <shtrb> Other than tlp what other power management elements tools might be loaded and could affect power state of ethernet devices (follow up on yesterday bug )
1152[14:13:59] <RadoS> jelly, I correct myself: it says processor 0-3 for both, but core-ids are different. Yet, still, why does it appear "same" in "top"?
1170[14:34:46] <RadoS> Hmm, so I don't understand enough about CPU specifications, but when the intro to both CPUs says 2 core 4 threads, the other 4 core 8 threads, this doesn't automatically convert to double cpu power?!
1184[14:41:42] <jelly> RadoS, desirable for what? If you want less power use, less cores is probably better. If you have highly parallelizable load, more cores is better
1185[14:41:48] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
1186[14:42:07] <shtrb> (*) Assuming your software actually use all of them properly
1208[14:48:37] <RadoS> jelly, processing power (for virtualization or high-end games). But if in the end only the total thread number makes a/ no difference, then core # doesn't matter either (except for power consumption).
1216[14:51:29] *** Quits: Souler (~Souler@replaced-ip) (Remote host closed the connection)
1217[14:51:35] <jelly> RadoS, assuming thermals are configured properly, unlike these Thinkpads, then more cores is better. SMT (HT) is a hack that never gives you as much processing power as full cores
1218[14:51:42] <RadoS> Hmm, maybe for virtualization cores/threads does make difference?
1242[15:09:56] <shtrb> jelly, wait wait , that could be close to my issue with power management with Ethernet ports, is there a chance that it could trigger Wifi (the dreaded I219-V kernel issue from 4.18 and up)
1243[15:10:19] <shtrb> sorry not wifi but ethernet card going into power save mode or something like that
1246[15:11:15] <jelly> shtrb, no idea, is it a laptop?
1247[15:12:10] <shtrb> yes , Thinkpad , a known firmware/driver bug where the I219-V when using e1000e would go to a non responsive state (the workaround is to either boot into windows or disable power mangment )
1309[16:01:05] <jelly> huma, only fixes for serious errors and security issues get applied
1310[16:02:07] <jelly> huma, if you're willing, you can check whether the problem is still present in the version in bullseye, the testing branch, if it's still there it's worth poking the maintainers again so it gets fixed for Debian 11
1448[18:09:42] <cluelessperson> there seem to be two directories with the same name...
1449[18:09:53] <cluelessperson> not sure how to pick a certain one
1450[18:10:54] <slowly_stuck> How do I create a virtual network interface that's not connected to any physical interface? (Apparently just defining it in /etc/network/interfaces doesn't work - ifup can't find the interface)
1470[18:27:13] <slowly_stuck> interesting, cluelessperson! I think a dummy isn't exactly right, since I'm next going to add it to a bridge (and use it to communicate with VMs that will also be on the bridge). gnat_x: is it reasonable to attach to an actual interface, but use a totally different network config?
1471[18:27:15] <neilthereildeil> whats the latest kernel shipping in debian 10?
1482[18:29:56] <greycat> You're using a backports kernel. "bpo" = backports.org or something like that.
1483[18:29:59] <neilthereildeil> im not sure. i might have pinned the kernel to that version, but my wifi isnt working correctly so i wanted to see if upgrading will help
1577[20:04:35] <greycat> Did you blindly change every "buster" to "bullseye" including third-party repos without actually CHECKING whether the third-party repo has a bullseye section?
1578[20:05:07] <ratrace> solomomi: first of all, bullseye-backports and bullseye-security do not exist yet. second, you should check docker docs if and how they provide repos for Bullseye
1579[20:05:40] <solomomi> retrace, I thought the docker are commented, isn't it?
1580[20:05:53] <ratrace> line 4 is not
1581[20:05:53] *** Quits: milkt (~debian@replaced-ip) (Remote host closed the connection)
1615[20:18:16] <ratrace> solomomi: jelly's advice is most sound. to upgrade, you should completely disable third party repos and then reinstall what got dropped, after upgrade. but first check if docker even provides bullseye repo
1616[20:18:41] <solomomi> ratrace it doesn't
1617[20:18:54] <solomomi> but i have a backup so I'm not afraid to mess it up :D
1618[20:19:03] <greycat> Oh, how cute. The command from ratrace's superuser URL doesn't "work" on google-chrome-stable because it has "stable" in its deb line.
1640[20:32:10] <ratrace> greycat: I don't understand that. isn't the most obvious thing, new libc, that requires relinking?
1641[20:32:57] <jken> Hello, I have a strange issue I am hoping someone here might know something about. Running buster, default kernel, I am using an USB to Ethernet adapter for network connectivity. If my router goes offline for a short period of time, network connectivity returns when it comes back online. HOWEVER, if the router loses power for a long period of time (confirmed at 6 hours), network connectivity never returns when the router is powered back up. When this
1642[20:32:57] <jken> happens, I don't even see my ehternet device in `ifconfig`
1650[20:34:04] *** Quits: tyzef (~tyzef@replaced-ip) (Quit: Live is not a bed of roses / La vie est un long fleuve tranquille)
1651[20:34:28] <jken> ratrace, is that disable-able?
1652[20:34:57] <ratrace> could be, I don't know how to even check for USB network devices
1653[20:35:33] <sigint> Hi, I built a new machine with an AMD GPU (5700 xt) using the amdgpu driver. X11 gives me screen tearing (and xserver crashes when enabling TearFree). Wayland is super laggy and consumes 100% CPU when moving the cusor quickly (but no tearing).
1654[20:36:00] <sigint> Which problem should I try to tackle? Tearing on X11 or laggy Wayland?
1663[20:37:55] *** Quits: Souler (~Souler@replaced-ip) (Remote host closed the connection)
1664[20:38:00] <ratrace> jken: if anyone here knows, they'll pipe up. you can ask again in a while if nobody does. on irc it's not uncommon to wait for days until you get an answer :)
1667[20:39:16] *** Quits: Souler (~Souler@replaced-ip) (Remote host closed the connection)
1668[20:39:21] <ratrace> sigint: "wayland is super laggy" ... wayland is a protocol. which wayland compositor are you using and is super laggy?
1669[20:39:33] <jken> Thanks ratrace!
1670[20:39:36] <enyc> Wonder where to find out about fixing cryptsetup-initramfs
1671[20:39:47] <slowly_stuck> jken: from some googling, do you have a /sys/bus/usb/devices/*/power/autosuspend? and if so, what's the content?
1672[20:39:57] <ratrace> enyc: just describe your problem
1673[20:40:15] <enyc> I have a mess where on debian/deriv, have lost working initramfs cryptsetup enabled with the right prompting for right uuid, to access encrypted rootfs.
1674[20:40:37] <ratrace> we don't do derivatives here
1675[20:42:04] <enyc> can mount the / and /boot and then -o rbind mount /dev /proc /sys /run into the chroot of 2:2.1.0-5+deb10u2 debian-normal cryptsetup-initramfs and update-grub, but in this chroot'ed-in state, it won't 'detect' what crypttab file or so needs creating inside initramfs to make it mount root, from what I can see.
1676[20:42:39] * enyc checks dd separating and un-cpio'ing a different deb10 cryptroot initramfs ..... cehcking for cryptroot file in there
1677[20:42:44] <sigint> ratrace, the default compositor that GNOME/gdm3 use I guess. I'm not sure how to check that
1678[20:42:46] <ratrace> initramfs tools require the /dev/mapper device to be exactly the SAME as the one listed in /etc/crypttab, on debian
1679[20:43:07] <ratrace> sigint: gnome then, yeah. mutter. then look up gnome bug tracker to see if anyone complained about the same. I'd also try a newer kernel from backports, see if that helps.
1682[20:43:53] <ratrace> sigint: that said.... super poor gnome performance under wayland was a known but, I just can't remember if it was fixed _before_ of after gnome version in buster. I think it's _after_ but don't call me on the witness stand for that :)
1707[20:53:32] <bryanpedini> Hello everyone! (happy new year if you haven't been already told so :))
1708[20:53:32] <bryanpedini> So, Dolphin 18.08.0 question (possibly bug), it doesn't move the favorite it should move, but the one before (replaced-url
1746[21:30:22] <jhutchins> diogenes_: It was a really solid system through 3.5, but when they re-wrote everything for 4.0 they made it way too bloated and complex and broke lots of things. There were at least three different databases trying to "index" everything.
1761[21:39:23] <prg3> Plasma is really sweet nowadays and if you like the old "shadow of the might it used to be" you can use trinity which is still maintained too
1765[21:41:11] <shtrb> prg3, it depends on what version you got :P
1766[21:41:18] *** Quits: fflori (~fflori@replaced-ip) (Remote host closed the connection)
1767[21:41:28] <prg3> Not sure what you are referencing about databases. Baloo indexes file metadata in a database - which probably makes sense. You can disable it if you want. Adokandai was dumb, yes, but plasma has moved away from it and i don't even think you need to have it installed anymore
1768[21:41:50] <prg3> true
1769[21:43:13] <ratrace> you forgot nepomuk
1770[21:43:15] <shtrb> jhutchins, I'm using plasma , and while 3.5 was perhaps the best UI/UX I had seen (with plasma) the recent versions in Debian are not that bad. it has some hickups (like breaking the notification system , and modifying configs like wtf people )
1771[21:43:17] <ratrace> the bane of performance under kde
1772[21:43:38] <jhutchins> One of the best things in 3.5 (which cleaned up a lot of bugs and bloat) was the integrated address book. Same address book for word processor and email and ponebook.
1773[21:43:38] <shtrb> ratrace, please leave this channel family friendly and do not use vile words
1774[21:43:45] <ratrace> mwahahahaha
1775[21:43:59] <shtrb> jhutchins, kaddressbook still exist (but now it is bundeled with akonadi)
1776[21:44:00] <jhutchins> Then 4 added 3000 blank lines to the databases, and it wouldn't even load any more.
1778[21:44:24] <shtrb> And you can use kaddresbook with just a file locally
1779[21:44:38] <jhutchins> That's something I like about Debian and xfce, I don't have to upgrade my hardware like you do with Apple. I can get years out of a functional system.
1780[21:44:42] <prg3> Oooh - reading into it now; that sounds awful
1781[21:44:46] <prg3> (nepomuk)
1782[21:45:53] * shtrb still remember when baloo used to index network drives by default and your home dir :D
1784[21:46:12] <prg3> To answer my earlier question, it looks like there are decent waterfox builds for debian here replaced-url
1785[21:46:53] <jhutchins> Yeah, that was one of the problems, trying to index the main storage array from an under-powered client workstation, in real time, while I was trying to work.
1787[21:47:14] <shtrb> At least Alex fixed that really fas
1788[21:47:16] <shtrb> *fast
1789[21:47:23] <prg3> Yeah baloo can be ... questionable to say the least heh
1790[21:47:32] <zutat> does gnome3 support umask yet?
1791[21:48:33] <bryanpedini> if there is one single thing that Windows does right (except being idiot AF and impossible to repair while easier to reinstall), is "office hours" activity... besides Windows Updates which does the hell it wants regardless, other things run only when the system is inactive
1796[21:49:47] <ratrace> windows update does that because users are morons who still, today, use windows7
1797[21:49:49] <shtrb> bryanpedini, I hope I had not offended you , but there is a slew of issues of Windows services that make your life hard as user
1798[21:50:16] <shtrb> It is so bad , that I disconnect it from the network to be able to work with it
1799[21:50:26] <bryanpedini> ye, basically everything else on that operating system (starting from the minecraft-look of the start menu icons and tiles and ending with the automatic reboot while *you're fricking working*) is a fricking total garbage! but at least they got "office hours" :)
1800[21:50:49] <bryanpedini> so... there's that :)
1802[21:51:26] <wiscii> yep, they extorted your working hours data out of you ..
1803[21:52:08] <zutat> shtrb: you forgot about textual ads in the terminal
1804[21:52:15] <bryanpedini> ratrace: of course users still use Windows 7... I'm about to install an SSD on my father's 9yo laptop, which Windows version do ya think I'm gonna install? that bloatware adware malware spy crap called 10? I'm not a moron, I'm privacy concerned...
1805[21:52:23] <shtrb> Oh and don't let me started about the AI #%@#% they do for payed hosted e-mails
1806[21:52:29] <shtrb> zutat, I don't get it (yet)
1807[21:52:54] <zutat> shtrb: it appears on powershell
1808[21:52:55] <shtrb> bryanpedini, install debian :P and rip all cloud services
1809[21:53:05] <shtrb> zutat, I guess I'm lucky as I don't get it (yet)
1810[21:53:08] <ratrace> bryanpedini: well good luck with soon-to-become windows7 bloatware malware fest crap as it gets infected becasue there's no longer security patches for it
1812[21:53:30] <bryanpedini> shtrb: if it was that easy to convert "normies" to Linux, I would live in a better world already
1813[21:53:35] <robobox> yeah, we *are* in the debian channel
1814[21:53:39] <shtrb> ratrace, you can alaways route windows traffic via a Debian VM :P
1815[21:53:40] <robobox> it is, actaully
1816[21:53:48] <ratrace> shtrb: as if that would help :)
1817[21:54:18] <shtrb> ratrace, ufw + proper rules would save your gluteus maximus most of the time
1818[21:54:20] <bryanpedini> ratrace: true what shtrb said about routing, but anyway I feel safer on a EOL Windows 7 under a serious Linux router/firewall unit, rather than having Windows 10 at all :)
1819[21:54:32] <robobox> especially if they were in an office in the 80s or 90s
1820[21:54:54] <robobox> normal people had to use unix then, so they should be able to use it
1821[21:55:06] <bryanpedini> even "naked" iptables is better than W10 integrated firewall + Win Defender together...
1822[21:55:07] <ratrace> shtrb: you're talking only bout a subset of vectors, via open ports. majority of infections come through trojans and bugs in, say, chrome
1823[21:55:08] <robobox> of course... there is the new generation that grew up on NT
1824[21:55:11] <shtrb> bryanpedini, people don't know Debian the same way they don't know Windows (IT) , give them the basics and they will find themselvs in a nice life
1825[21:55:26] <shtrb> I had a ~90 year old person using debian + plasma without too many problems few years ago
1826[21:56:03] <bryanpedini> shtrb: absolutely true, tho normies are lazy guys on their couches watching TV and don't want to even worry about learning what they already know
1827[21:56:29] <shtrb> Here's the point , many people don't know windows either
1828[21:56:31] <robobox> that may be true for youngsters
1829[21:56:38] <bryanpedini> my grandpa hated when I converted his still-working (1yo) WinXP to 7, let alone if I shown him 10, *or Deb!!*
1830[21:56:39] <shtrb> feces , I work with debian and I don't know it either
1831[21:56:40] <robobox> but not for older peeps
1832[21:57:03] <wsky> using and bootstraping are two different things
1833[21:57:08] <shtrb> If not ratrace I would be still getting apparmor messages
1840[21:58:15] <bryanpedini> let me try this: I'll put Debian + Mate and I'm gonna change start menu icon and background to WinXP, then pretend it was WinXP like nothing has ever happened :) :) :)
1867[22:01:14] <wsky> so they were forced to take that down
1868[22:01:21] <bryanpedini> wsky: didn't they argue they created themselves just "taking inspiration" from Win? :P
1869[22:01:38] <wsky> no, they were literarly icons taken out from xp
1870[22:01:44] <bryanpedini> upsie...
1871[22:02:05] <wsky> with that it was possible to make you linux to look very alike xp
1872[22:02:10] <jelly> bryanpedini, new edge is just chromium with a different sync backend and default search engine, not much porting to do
1873[22:02:11] <wsky> your linux*
1874[22:02:13] <bryanpedini> better have added that one pixel with a slight color change so MD5 would vary and YT algorithm wouldn't have picked it up...
1876[22:03:38] * jhutchins checks the /topic. Yup, Debian support, not Windows politics.
1877[22:03:38] <bryanpedini> jelly: and that's why I understand even less the needing of a Microsoft-branded browser if it's just Chromium with MS logo and different search engine on top, why would I want to constantly be reminded what stupid company created the OS I'm running when browsing the web, and most importantly search over Bing? (seriously?)
1897[22:09:15] * shtrb think ratrace was just testing all of us
1898[22:10:51] <ratrace> shtrb: looks like "linux" it is. nope, I wasn't testing. I'm really dumbfounded how complex and unintuitive this crap is. And then they want us to contribute. HAHAHAHA.
1900[22:11:28] <jelly> bryanpedini, don't ignore the sync backend, using a single account for all tools bundled with the OS is convenient
1901[22:11:59] <bryanpedini> ratrace: well, actually it is simple, just download from kernel.org and contribute (than it's a matter of the complexity that there is behind which damn git system/server do they use, but that's another story)
1902[22:12:17] <shtrb> git is easier than quilt
1903[22:12:18] <ratrace> the bug I'm trying to fix is specific to debian patching
1908[22:13:44] <ratrace> I don't need the upstream kernel, and I know how to build them, been a gentoo user longer than debian. I needed specifically the kernel source package for debian, but here's a multitude of them and tryig what you ordinarily would with apt-get source <regular package name> did not work
1909[22:13:45] <sney> ratrace: maybe there are some clues here and in the repo generally, replaced-url
1910[22:13:51] <bryanpedini> shtrb: never heard of quilt, anyway I was refering about the complexity behind setup and usage of GitHub/GitLab/Google Code server/etc etc
1911[22:14:24] <ratrace> sney: what's worse, even this is wrong: replaced-url
1912[22:14:45] <ratrace> but nvm... shtrb was on it, "linux" is the source package name
1989[23:23:36] <wisbit> good evening guys, I have created a service containing a /usr/bin/java -jar /home/user/App/app.jar --headless </dev/null &>/home/user/App/app.log
1990[23:23:51] <wisbit> the problem is that it doesn't send the output in that log file
1991[23:24:20] <wisbit> while, starting that command manually works well and the app.log is correctly used
1992[23:24:24] <ratrace> wisbit: that's on the Exec line of the .service unit?
1993[23:24:56] <wisbit> ratrace, that's correct
1994[23:24:59] <ratrace> yeah, no. the Exec= line is not a shell.... < > & redirection doesn't work there.
1995[23:25:06] <greycat> 1) you can't use shell redirections in Exec= at all; 2) &> is not a standard shell redirection; it's bash crap.
1996[23:25:27] <wisbit> fair enough, it is indeed used in bash
1997[23:25:33] <greycat> 3) systemd units can't log to a human-readable file at all.
1998[23:25:43] <rk4> :(
1999[23:25:48] <ratrace> actually they can
2000[23:26:04] <greycat> if they can, they sure went out of their way to hide that
2001[23:26:44] <ratrace> StandardOutput= and StandardError= directives, see systemd.exec(5)
2025[23:34:18] <ratrace> I'm gonna hazard a guess and say: no.
2026[23:34:47] <ddsys> it was enabled by default
2027[23:35:07] <ddsys> xfce
2028[23:35:10] <greycat> the obvious answer is "disable it and see if anything breaks"
2029[23:35:21] <ddsys> right
2030[23:36:08] <BCMM> ddsys: you've probably installed a desktop task that includes accessibility tools
2031[23:36:23] <ddsys> hmm
2032[23:36:51] <BCMM> removing it may remove your desktop task, so be careful with autoremove afterwards
2033[23:36:53] <n4dir> pacpl, a perl tool to convert audio files from one format to another, perl complaints but syntax warnings (errors in the future, if i understand correct). Anyone uses it and has a workaround? replaced-url
2034[23:37:01] <n4dir> same for pastebinit and python, btw
2035[23:37:28] <n4dir> "just ignore" is the answer i am looking for
2036[23:37:37] <ddsys> BCMM: thanks
2037[23:37:41] <greycat> has a bug already been filed for those warnings?
2038[23:38:01] *** debhelper sets mode: +l 1198
2039[23:38:02] <n4dir> greycat: damnit. let me check. I wanted to, then got confused. sorry
2046[23:40:04] <greycat> Mon Jan 4 17:39:58 EST 2021
2047[23:40:04] <greycat> Mon 04 Jan 2021 05:39:58 PM EST
2048[23:40:16] <Elodin> i want the portuguese_brazilian format
2049[23:40:25] <greycat> OK, then set it however you like ;-)
2050[23:40:27] <Elodin> where i see this stuff?
2051[23:40:32] <greycat> !locales
2052[23:40:32] <dpkg> Use 'dpkg-reconfigure locales' to get it up and running. This generates <locale> definitions and also edits /etc/default/locale which sets the $LANG environment variable at login time. Use "LANG=C command" to change the output language for a one off command, ask me about <localised errors>. See also <mac locales>. replaced-url
2079[23:52:06] *** Quits: rgr (~rgr@replaced-ip) (Remote host closed the connection)
2080[23:52:53] <incal> hi, I just disabled every single systemd service, except getty@.service, without it Emacs wouldn't start in tty1 for some reason, but strangely, the other ttys with tmux were fine. the number of processes seems largely unaffected, internet is up as you can see. computer seems faster in general tho. I wonder what the implications are?